Do people regard ethnic outgroups as more pathogen-wise dangerous to interact with?

– by Lei Fan

Most people would rather touch flowers than wet rags, eat fresh foods than rotten ones, and drink water from one’s own bottle than from a stranger’s. These preferences seem to have straightforward functions: they protect us from the microbes that cause infectious diseases. Researchers suggest that we humans – like many other animals – have evolved some ability to detect and avoid pathogens – that is, we have a behavioral immune system.

A popular hypothesis in the behavioral immune system literature suggests that negative attitudes toward people with different skin color, ethnical origins, or religions – that is, outgroup members – serve pathogen-neutralizing functions. The line of thinking goes as follows: different groups live in different areas, and those areas have different pathogens. One group carries pathogens that they have acquired some immunity against, but that another group has not acquired immunity against. Hence, the average outgroup individual is more likely to carry novel pathogens than the average ingroup individual is, which makes these intergroup interactions pose higher infectious disease threats. Historical events illustrate the most extreme disease consequences of such interactions. For example, the introduction of measles and smallpox from Europeans to Native Americans led to a 57% reduction in effective population size (Lindo et al., 2016). Consistent with these ideas, many studies have reported that intergroup biases are stronger in areas with higher pathogen threats (e.g., Letendre et al., 2010, O’Shea et al., 2019), after research participants have been primed to feel vulnerable to disease (e.g., Faulkner et al., 2004, Klavina et al., 2011), and among individuals who report greater motivations to avoid infection (e.g., Aarøe et al., 2017; Clifford et al., 2022). However, other findings haven’t been straightforwardly consistent with the idea that outgroups are, in general, interpreted as pathogen threats (e.g., Ji et al., 2019, Karinen et al., 2019, van Leeuwen and Petersen, 2018).

Inspired by the test that van Leeuwen and Petersen (2018) conducted, we aimed to test a foundational aspect of the hypothesis that the behavioral immune system interprets outgroup members as more of a pathogen threat than ingroup members: that people should feel less comfortable with microbe-sharing contact with ethnic outgroup targets than ethnic ingroup targets.

This paper was a registered report – a format in which our hypotheses, methods, and analysis plan were peer-reviewed before we collected the data and knew the study’s outcomes. We recruited participants who were either White UK residents or East-Asian Chinese residents. Each participant saw one of 40 faces, which were either East Asian or White. Some participants saw a face that was unmodified; others saw a face modified to have an infection symptom (shingles); and others saw a face modified to be wearing a facemask. Participants completed 7 items in which they reported how comfortable they would be with indirect contacts with the target, such as using drinking from the same bottle as the target, and sitting next to the target while the target was coughing and sneezing.

Our results revealed that contact comfort was lower for targets modified to have symptoms of infection relative to targets that didn’t have these symptoms. And people who scored higher on an instrument assessing the tendency to be disgusted by things that contain pathogens (e.g., dog feces) reported lower contact comfort. So, we had some evidence that the contact comfort items worked as intended. But were people less comfortable with contact with outgroup members than ingroup members? No – there was no interaction between participant ethnicity and target ethnicity. The presence or absence of infection cues (or a facemask) did not change this (lack of) effect.

Hence, do we interpret ethnic (outgroup) membership as diagnostic of infectious disease risk? Results of the current study suggest that no, we do not. But other results from previous studies seem still providing a lot of evidence supporting the idea that the behavioral immune system relates to intergroup attitudes in some way, such as those mentioned earlier. Thus, the main take-home message from our paper is, the story behind these mechanisms is probably way more complex than group membership being treated as solely indicative of pathogen threat.